Section 5: Types of Orbits, or Why Satellites Are Where They Are

The choice of a particular orbit for a satellite depends mainly on its mission. For example, a remote-sensing satellite that collects high-resolution images of the Earths surface should be as close to the Earth as practical. Consequently, such satellites are in low earth orbits. A commercial broadcast or communications satellite has other requirements. It should be able to send and receive signals from a large geographic area. It should preferably be in a fixed location, so ground stations will not need expensive satellite-tracking equipment. For these reasons, most communications satellites are in equatorial geostationary orbits. The orbits for other satellites will similarly be chosen based on their missions. This tight correlation between mission and orbit has important consequences. For example, although the traditional notion of territory as a fixed ground area or fixed volume of airspace is not relevant in space (since all permanent residents must be orbiting), a few special orbits are uniquely suited to a specific purpose and are therefore highly valuable. As a result, it is often possible to guess at the function of an unknown satellite by observing what orbit it follows. In this section, we first discuss several important characteristics of orbits. These include those proscribed by geometry: the motion of satellites with respect to the Earth, the elevation angle above the horizon of the satellite for different positions on the ground, the maximum ground area that a satellite can observe, and the time it takes for a transmission signal to travel between the ground and satellite. Other orbital characteristics are a consequence of the local environment, such as radiation and atmospheric effects. We then discuss the most common types of orbits and the satellites that populate them.
T H E C O N ST RA I N TS O F G E O M ET RY

The Motion of Satellites Relative to the Surface of the Earth

In addition to the orbital motion of the satellite, the Earth also spins on its axis, and the motion of the satellite relative to the surface of the Earth is determined by both effects. Because of the rotational motion of the Earth, when a satellite returns to the same point in its orbit one period later, it is no longer over the same location on the Earthunless the orbit is chosen so that the satellites period is one day. In this case, the satellites orbital period is the same as the Earths rotational period, and such orbits are called geosynchronous. Geosynchronous orbits can be circular or elliptical and can have any inclination angle. As shown in Section 4 (see Figure 4.2 and Table 4.2), circular orbits with an altitude of about 36,000 km are geosynchronous.

T Y P E S O F O R B I T S , O R W H Y S AT E L L I T E S A R E W H E R E T H E Y A R E

29

A circular geosynchronous orbit that lies in the equatorial plane (inclination of 0) is a special case: it is geostationary. A satellite in this orbit stays fixed relative to the surface of the Earth and remains directly over a point on the equator. To an observer on the ground, the satellite appears motionless. Note that only in an equatorial orbit is it possible for a satellite to remain stationary over a point. Because there is only one geostationary orbit, space in this orbit is valuable. Now consider the case of a satellite in orbit at several hundred kilometers altitude. As shown in Figure 4.2 and Table 4.2, such satellites have orbital periods of roughly 90 minutes. In 90 minutes, the surface of the Earth at the equator rotates about 2,500 km (at other latitudes, the distance rotated would be less than 2,500 kilometers). Thus, after one period, the satellite passes over the equator at a spot 2,500 km west of the spot it passed over on its previous orbit. A satellite in a low-altitude orbit also is in view of a given location for only a short time. To a person on the ground directly under the orbit, the satellite appears above the horizon on one side of the sky, crosses the sky, and disappears beyond the opposite horizon in about 10 minutes. It reappears after about 80 minutes,1 but does not pass directly overhead (unless the observer is at one of the poles), since the Earth has rotated during that time. As the satellite moves in its orbit, the point on the Earth directly beneath the satellite traces out a path called the satellites ground track. (The ground track of a geostationary satellite is simply a point on the equator.)2 Figure 5.1 shows an orbit with inclination of 45; the shaded disc is that part of the orbital plane lying inside the orbit. The line where the orbital plane touches the Earths surface would be the ground track of the satellite if the Earth did not rotate. This figure shows why the ground track, when drawn on a flat map of the Earth, appears as a curve that passes above and below the equator, as shown in the upper panel of Figure 5.2. Half of the orbit lies above the equatorial plane and half lies below. Note that the maximum latitude the ground track reaches (north and south) is equal to the inclination of the orbit. Because the Earth rotates, the ground track does not lie in the same place on the Earths surface during its next orbit, as shown by the dashed line in the bottom panel of Figure 5.2. Unless the period of the satellite is chosen to have a special value, the satellite in time flies over all points of the Earth between the maximum and minimum latitudes.

1. If it does not pass overhead on the subsequent orbit, it will not be visible unless it is at high enough altitude. 2. For a satellite in a geosynchronous orbit that is not geostationary, with a nonzero inclination , the ground track will be a figure eight centered at a fixed point on the equator. The top and bottom of the 8 will lie at latitude above and below the equator.

30

T H E P H Y S I C S O F S PA C E S E C U R I T Y

Figure 5.1. The shaded disc is that portion of the orbital plane lying within an inclined circular orbit. If the Earth did not rotate, the line where the shaded disc meets the Earths surface would be the ground track of the satellite. This figure shows that, for an inclined orbit, the ground track of the satellite passes above and below the equator.

Figure 5.2. The upper panel shows the ground track for one pass of a satellite in a low earth orbit with an inclination angle of 45. The track of the satellite on its next pass over this region is shown as the dashed curve in the lower panel. Since the Earth has rotated in the time the satellite was making one orbit, the ground tracks do not overlap.

T Y P E S O F O R B I T S , O R W H Y S AT E L L I T E S A R E W H E R E T H E Y A R E

31

Elevation Angle of the Satellite

The elevation angle of a satellite is the angle between the satellite and the local horizon as seen from a particular location on the ground (Figure 5.3). It is a measure of how directly overhead a satellite is at a given time, with an elevation of 90 signifying the satellite is directly overhead. Because it is measured with respect to a specific ground location, it is different for different observers on the ground. It varies with time as the satellite moves through its orbit.
Figure 5.3. The elevation angle of a satellite at a given time for an observer at the point P on the Earth is the angle between the local horizon at P and a line from the observer to the satellite.

Specifically, the elevation angle at a given time depends on several parameters that describe the relative location of the satellite and the observer on the ground. These include the latitude and longitude of the observer, the altitude of the satellite, the inclination angle of the orbit, and where on the orbit the satellite is (the satellites latitude and longitude). The exact relation is given in the Appendix to Section 5. A few examples can provide a sense of how the observers latitude, satellite altitude, and orbital inclination affect the elevation angle. Consider an observer at the equator (i.e., at a latitude of 0) and a satellite in a circular equatorial orbit (i.e., at an inclination angle of 0). As the satellite traverses its orbit, it passes directly over the observer, and its elevation angle increases from 0 to 90 and then decrease back to 0 for an observer at any point on the equator. The satellite has an elevation angle of 90 only for points on the ground directly beneath its orbit, so for observers not at the equator, the satellite never appears directly overhead. Instead, its maximum elevation angle depends on the observers latitude, the altitude of the satellite orbit when it has the same longitude as the observer, and the altitude of the orbit at apogee and perigee

32

T H E P H Y S I C S O F S PA C E S E C U R I T Y

(see the Appendix to Section 5). For example, for an observer at latitude 45 and a satellite in a circular equatorial orbit with an altitude of 500 km, the maximum elevation angle will be only 17. The maximum elevation angle increases with satellite altitude: for a satellite in geostationary orbit at an altitude of 36,000 km, the highest elevation angle seen by the observer at 45 latitude is 38, which occurs when the satellite and the observer are at the same longitude. Because the Earth rotates, a satellite regularly passes directly over parts of the Earth with latitude equal to or smaller than the angle of the satellites inclination. For areas of the Earth at higher latitudes, the satellite may be observable but it will never be overhead; its maximum elevation angle will be less than 90. So a satellite in an orbit with an inclination angle of 45 will have a maximum elevation angle of 90 for all points of the Earth between 45 north and 45 south. A satellite in a polar orbit passes directly overhead all points on Earth. Two types of satellites pass directly over the same area of the Earth on each repetition of their orbit: a satellite in an equatorial orbit at any altitude and a satellite in a geosynchronous orbit. At other points on the Earth, these satellites never appear directly overhead. The elevation angle of a satellite at a given location has a strong impact on how it can be used, and thus can suggest its purpose. For example, a ground station has a difficult time receiving signals from a satellite that is at a low elevation, for two reasons. First, a signal from the satellite travels a longer distance through dense atmosphere than it would if sent from a satellite at a higher elevation angle, which results in a greater attenuation of the signal strength. Second, objects on the horizonsuch as tall buildings or mountainsmay be in the line of sight between the ground station and satellite, thereby blocking the signal transmission. In densely built cities, tall buildings can block signals sent to and from a station on the ground for low elevation angles, in the worst cases even up to 70, so many satellite receivers and transmitters in cities are mounted on the tops of buildings. For some applications, however, the receivers need to be mobile and on the ground, such as ground vehicles using the Navstar Global Positioning System (GPS). GPS satellites operate at 55 inclination and may not be able to adequately serve all their potential users in urban settings. Japan, for example, lies at 30 to 45 latitude so the GPS satellites never pass overhead, and it has high buildings in its urban areas and high mountains in its rural areas that can block GPS signals. It is developing a system of three satellites in highly elliptical orbits with their apogees over Asia (Quazi-Zenith Satellite System (QZSS)). These will work with the US GPS system to better serve Japans population. Geostationary communications satellites are also less useful to Russia than to the United States: equatorial orbits do not afford good coverage of the poles or regions at very high latitudes, and many key Russian military installations are in the north polar region. Instead, Russia uses satellites in highly inclined orbits that can be easily seen from northern latitudes during part of their orbit. When these orbits are highly elliptical and have their apogee near the North Pole, the satellites in these orbits appear overhead for longer periods of time, making them particularly useful. Orbits of this type include Molniya orbits, which are discussed below.

T Y P E S O F O R B I T S , O R W H Y S AT E L L I T E S A R E W H E R E T H E Y A R E

33

Observable Ground Area

How much of the Earths surface can be observed from a satellite depends on its altitude h. While the altitude determines the maximum area the satellite can observe, the actual observable area may be limited by the sensors the satellite carries, which may not be able to view this entire area simultaneously. The outer edge of the observable region is a circle, the radius of which depends only on the satellites altitude. The relation between these two parameters is given in the Appendix to Section 5, and illustrated in Figure 5.4. However, a ground station can generally communicate with a satellite only if it is at an elevation angle greater than some minimum value; this minimum value is typically 5 to 10. As a result, the effective ground area with which the satellite can communicate is less than the full area the satellite can observe. The radius of the effective region is a function of both satellite altitude h and the minimum elevation angle, as discussed in the Appendix to Section 5.
Figure 5.4. The size of the area of the Earth observable from a satellite depends on its orbital altitude. The observable area is compared here for satellites at two different altitudes: the satellite at the lower altitude sees a much smaller area than the one at the higher altitude. Note that the observable area also describes the area on the Earth that can see the satellite.

Table 5.1 lists the radius of the maximum observable ground area by satellites at several altitudes, and the radius of the effective ground area when the minimum elevation angle is 10. Note that for satellites at low altitudes the effective area is roughly half of the maximum observable area, while for

34

T H E P H Y S I C S O F S PA C E S E C U R I T Y

higher orbits the effective area is not reduced as much relative to the maximum area. Note also the much larger fraction of the Earth that is visible from a satellite in geosynchronous orbit compared with low earth orbit.

Table 5.1. This table shows the radius (as measured along the Earths surface) of the maximum region of the Earth that satellites at several altitudes can see, as well as the percentage of the Earths surface that region covers. It also shows the size of the effective observable area if the minimum elevation angle at which the ground station can communicate with the satellite is 10. Effective observable region (Minimum elevation Maximum observable region Radius Satellite altitude (kilometers) 500 1,000 20,000 (Semisynchronous) 36,000 (Geosynchronous) (km) 2440 3360 8450 9040 % of Earths surface 3.6 6.8 38 42 angle = 10 degrees) Radius (km) 1560 2440 7360 7950 % of Earths surface 1.5 3.6 30 34

Clearly, the higher the orbit, the larger the ground area that the satellite can observe and communicate with. However, other factors also affect the choice of satellite altitude. The intensity of electromagnetic radiationincluding visible light, infrared, and radio wavesdrops off in proportion to the square of the distance between the sender and receiver. This drop in signal intensity with increasing altitude suggests that a lower orbit is preferable. On the other hand, for a given coverage area on the ground, a lower orbit satellite must propagate the signal over a wider angle than a higher orbit satellite would (see Figure 5.4). To achieve this wide dispersion, the satellite antenna sacrifices signal strength (or gain) in any single direction, and this partially offsets the distance advantage. Thus, for a given satellite mission, the tradeoffs lie between a low-altitude satellite with a wide-area antenna and a high-altitude satellite with a highly directional antenna. The satellites observable ground area, as well as its motion with respect to the Earth, has important consequences for how it can be used. For example, a satellite taking high-resolution photographs of the ground is best placed in a low-altitude orbit. In this case, it will spend only a small fraction of its time in view of any particular location on the Earth. Constant low-altitude surveillance of a specific area would therefore require multiple satellites in orbit, so that one moves into position as another moves out of position. The total number of satellites required in a constellation (i.e., a system of more than one satellite) in order to have one satellite in the right place at all times is the absentee ratio, which depends on the ground area that each satellite covers. All else being equal, placing the satellites at higher altitudes

T Y P E S O F O R B I T S , O R W H Y S AT E L L I T E S A R E W H E R E T H E Y A R E

35

decreases the absentee ratio. However, using satellites at higher altitudes may be impossible for some applications, such as high-resolution surveillance or ballistic missile defense, which requires space-based interceptors to be relatively close to attacking missiles.

Transmission Time
The round-trip transmission time between a ground station and a satellite is the distance traveled divided by the speed of light (300,000 km/sec). The exact distance traveled depends on the elevation angle of the satellite and where it is in its orbit, but this is roughly twice the orbital height h of the satellite. The round-trip transmission time in seconds is roughly (2 h)/300,000, where h is expressed in kilometers. For a satellite in geosynchronous orbit at an altitude of 36,000 km, the round-trip transmission time is roughly 0.25 seconds. Because of this time delay, using such a satellite to relay data between two or more ground stations in its field of view requires echo control on telephone transmissions and special protocols for data transmission. In contrast, a satellite at an altitude of 500 km has a round-trip transmission time of only 0.003 seconds, eliminating the need for echo control or other special treatment.
EFFECTS OF THE LOCAL ENVIRONMENT

Interference
If neighboring satellites use the same transmission frequencies, a receiver on the ground may find that their transmitted signals interfere with one another if the satellites are too close together. Because geostationary orbits are a limited resource, satellites in this orbit are positioned close together and their transmission frequencies must be planned carefully so that their transmitted signals do not interfere with each other. The International Telecommunications Union (ITU) performs the task of assigning locations and frequencies to satellites in geostationary orbit. This is facilitated by the fact that these satellites remain in fixed positions with respect to each other. When this arrangement is not respected or the orbit becomes too crowded, interference problems have occurred.3 Satellites in different orbits are not stationary with respect to each other, so they pose a more difficult interference problem if they communicate at the same frequencies.4 Recently, the ITU designed regu-

3. In 1996, the United Nations reported that severe crowding in the geostationary orbital slots over Asia led to the jamming of a communication satellite by PT Pasifik Satellite Nusantara (PSN) of Jakarta, Indonesia, in defense of an orbital position claimed by Indonesia. This incident focused global attention on the worsening problem of orbital crowding and caused the matter to be brought before the 1997 World Radio Communication Conference (WRC) of the 187 member-nation ITU in Geneva (United Nations, Highlights in Space: Progress in Space Science, Technology, Applications, International Cooperation and Space Law, 1997 [Vienna: United Nations, 1998], 51). 4. The Skybridge low-earth-orbiting satellite system (www.skybridgesatellite.com, accessed January 15, 2005) operates in the Ku band, as do many geostationary satellites. To avoid interference, the Skybridge system turns off its transponders as the satellite approaches the

36

T H E P H Y S I C S O F S PA C E S E C U R I T Y

lations that allow satellites in low earth orbits to use the same frequencies as those in geostationary orbits by setting an upper limit to the amount of power they may broadcast.5

Radiation Environment
Space is a harsh environment.6 Satellites do not enjoy the protection of the atmosphere against radiation and particles from the Sun and the larger universe. This radiation environment and its changes are sometimes referred to as space weather. Solar ultraviolet and X-rays generally do not penetrate the skin of a satellite. Only during solar flares do X-rays have sufficient energy to penetrate a few millimeters of aluminum, but generally the frequency and duration of this radiation is not sufficient to damage internal electronics. These X-rays do, however, damage and degrade solar panels, which many satellites rely on to generate energy. Charged particles (positively charged protons and negatively charged electrons) are an additional concern. These particles can have very high energies and can damage and degrade electronics. These particles are trapped by the Earths magnetic field, forming two toroidal (doughnut-shaped) regions around the Earths equator known as the Van Allen belts (see Figure 5.5). The inner torus extends from an altitude of roughly 500 km to 5,500 km, with the highest particle density in the middle at about 3,000 km.7 The particle density is greatest at the equator and low latitudes, then decreases as the latitude increases. By latitude 50 or 60 north or south, the density in the belt is very low. This inner belt is populated primarily with high-energy protons that can readily penetrate spacecraft and, with prolonged exposure, damage instruments and endanger astronauts. (This region also contains highenergy electrons, but at a much lower density than the protons.) Both manned and unmanned spaceflights stay out of the high-density regions. The outer torus extends from an altitude of roughly 12,000 to 22,000 km, with the highest particle density also in the middle, at 15,000 to 20,000 km. The outer belt is populated primarily with high-energy electrons. The electron

equatorial plane where the geostationary satellites are. Skybridge needs to have more than one satellite in view at any one time to avoid interruption of service, adding complexity and expense to the system. See Kristi Coale, Small Satellites Push for Elbow Room, Wired News, October 14, 1997, http://www.wired.com/news/technology/0,1282,7657,00.html?tw=wn_story _mailer, accessed January 15, 2005. 5. The ITU amended the radio regulations at the World Radiocommunication Conference in 2000. The documentation pertaining to the regulation can be obtained from the ITU website http://www.itu.int, accessed January 15, 2005. 6. See The Earths Trapped Radiation Belts, NASA Space Vehicle Design Criteria (Environment), NASA SP-8116, (Springfield, Virginia: March 1975), available from the National Technical Information Service; NASA JPL Radiation Effects Group, Space Radiation Effects on Microelectronics, http://parts.jpl.nasa.gov/docs/Radcrs_Final.pdf, accessed January 15, 2005. 7. The belts will reach down to lower altitudes near the poles, as shown in Figure 5.5, but the particle density is low at that point.

T Y P E S O F O R B I T S , O R W H Y S AT E L L I T E S A R E W H E R E T H E Y A R E

37

distribution is much more variable spatially and temporally than the proton distribution. However, like the proton flux, the electron flux is highest near the equator and becomes negligible at a latitude of 60 north or south.

Figure 5.5. An illustration of the Van Allen radiation belts, 8 showing the inner radiation belts, which have a maximum density at an altitude of about 3,000 km and are primarily protons, and the outer belts, which have a maximum density at about 17,000 km and are primarily electrons. The solar wind (consisting mainly of low energy protons and electrons) distorts the toroidal shape and flattens the torus on the sunward side and creates a tail on the shaded side, which is shown by the solid lines on the right.

A region between the inner and outer belts, known as the slot, has a low density of high-energy electrons. This region extends from roughly 6,000 to 12,000 km, but can disappear during active solar periods when the inner and outer belts sometimes overlap. Because the radiation belts are well characterized, they can be accounted for in satellite design. But radiation shielding adds mass and expense. A few centimeters of aluminum are sufficient to stop most electrons. However, as the electrons entering the aluminum slow down, they can generate X-rays. Thus satellites in the most intense radiation environments may require heavy lead shielding to protect against these X-rays. In addition, extraordinary events can significantly change the radiation environment and endanger satellites not designed to withstand higher fluxes. Five or ten times a year, solar flares generate bursts of high-energy protons and electrons, which may occasionally be stronger than expected.

8. Adapted from NASA JPL Radiation Effects Group, Space Radiation Effects on Microelectronics, presentation, 2002, http://parts.jpl.nasa.gov/docs/Radcrs_Final.pdf, accessed January 15, 2005.

38

T H E P H Y S I C S O F S PA C E S E C U R I T Y

Thus the choice of orbit determines the amount of shielding a satellite needs, based on whether and what part of the Van Allen belts they will encounter. For example, at altitudes of less than 500 to 1,000 km, the density of charged particles is low and little shielding is required. These are the altitudes at which many satellites, and all extended missions with personnel, operate. On the other hand, the satellites of the U.S. Global Positioning System operate at an altitude of 20,000 kmwhere the density of highenergy electrons is at or near its maximum. This demonstrates the feasibility of shielding against even high concentrations of charged particles. Human actions can alter the radiation environment as well. In 1962, the United States conducted a nuclear test explosion called Starfish at an altitude of 400 km. It generated a significant perturbation in the trapped electron distribution. The maximum intensity of the perturbation was in the region over the equator at altitudes between 1,600 and 6,300 km, but effects reached out to at least 44,000 km. Below 1,600 km, the perturbation decayed fairly rapidly, but at these altitudes, the radiation level was still an order of magnitude higher than normal several years after the test.

Effects of the Atmosphere

There is no outer edge to the atmosphere. The air that makes up the atmosphere is held to the Earth by gravity, just as the water in the oceans is. And just as the water pressure increases with depth in the ocean, the atmosphere is most dense at ground level and thins out quickly with increasing altitude, falling off roughly exponentially. At 10 km altitude (the height of Mount Everest) the air is nearly too thin to breathe, and the density is about onethird of the density at sea level. At 100 km altitude, the density has dropped to less than one-millionth of that at sea level; by 600 km it is reduced by another factor of one million. For many purposes, the sensible atmosphere ends around 100 km, and this is generally accepted as the altitude at which space begins. However, for some purposes the effects of the atmosphere must be considered at altitudes higher than 100 km. For example, the atmospheric drag on satellites may be very small at altitudes of several hundred kilometers, but its cumulative effect over many orbits is not negligible. The atmosphere has important consequences for satellites in low orbits. One consequence of the balance of forces (centrifugal and gravity) that keep a satellite in orbit is that if atmospheric drag begins to slow a low-orbit satellite, it will no longer be moving fast enough to stay in its original orbit and will begin to spiral down toward the Earth. The increase of atmospheric drag at lower altitudes restricts satellite orbits to altitudes of a few hundred kilometers or higher. Satellites in low orbits must carry stationkeeping fuel so they can occasionally maneuver to offset the effects of atmospheric drag and stay in orbit. Moreover, because the atmospheric density varies spatially and temporally, there is an inherent limit to the accuracy with which the future position of a satellite in low orbit can be predicted. This fact is important for long-term monitoring of satellites.

T Y P E S O F O R B I T S , O R W H Y S AT E L L I T E S A R E W H E R E T H E Y A R E

39

More Complicated Effects of Gravity

The circular and elliptical orbits described in Section 4 considered only two forces on the satellite: Earths gravity and centrifugal force. In these cases, Earths gravity is assumed to be spherically symmetrical. However, because the Earth is not a perfect sphere, its gravitational field is not perfectly symmetric and this affects the orbital shape and orientationand changes them over timein subtle but important ways. The gravitational pull of the Sun, which is much weaker than that of the Earth, also affects the orbits. In the absence of these gravitational irregularities, the orientation of the orbital plane of a satellite would remain constant in space. However, because of them, the orbital plane precesses, that is, it rotates with respect to the Earths axis. The rate of precession depends on the eccentricity, altitude, and inclination of the orbit. The proper choice of these parameters can allow a satellite to use the precession rate to aid a specific purpose. For example, by choosing an orbit that precesses at a specific rate, the orbital plane can be made to keep a constant angle with respect to a line between the Earth and Sun throughout the year, as the Earth travels around the Sun. A satellite in such an orbita sun-synchronous orbitobserves each place on Earth at the same local time and sun angle; more detail on this orbit appears below. Gravitational irregularities also cause the major axis of an elliptical orbit to rotate slowly in its orbital plane. Thus, for an elliptical orbit inclined with respect to the equator, its apogee moves slowly from over one hemisphere (i.e., northern or southern) to over the other, then returns over the first. The rate at which this occurs depends on the inclination angle of the orbit; for an angle of 63.4, the rate is zero. Thus, the apogee of an orbit with this inclination angle remains over the same hemisphere. In fact, while precession still causes the apogee to slowly rotate about the Earths axis, it remains over the same latitude as it rotates. These and other orbits are discussed in more detail below.
COMMON ORBITS

Low Earth Orbits

Satellites in low earth orbits (LEO) operate at altitudes of hundreds of kilometers up to around 1,000 km. (Satellites at orbital heights of a few thousand kilometers could also be said to be in low earth orbits, but few satellites populate this part of space because of the large amount of radiation there.) LEO satellites have orbital periods of roughly 90 minutes. As noted above, space at these altitudes is mostly free from high radiation and charged particles. Atmospheric drag is small above a few hundred kilometers, although increases in solar activity can cause the outer layer of the Earths atmosphere to expand, increasing the drag on satellites in orbits in the lower part of this region. Since a satellite in LEO cannot see a large ground area and since it moves relative to the Earths surface, LEO may not seem to be useful for missions such as communications. However, a network that contains enough LEO satellites to see all regions of the Earth and that can relay signals between the

40

T H E P H Y S I C S O F S PA C E S E C U R I T Y

satellites can provide continuous worldwide coverage. If such a network includes polar or near-polar orbits, it can also provide coverage of polar and high latitude regions, as geostationary satellites cannot. Because they are in low orbits, the round-trip transmission time from these satellites is relatively short (0.005 seconds to and from the ground), eliminating the need for echo control or other special treatment. (The time required for signals transmitted over long distances around the Earth when relayed through multiple satellites is dominated by the distance along the Earth rather than the altitude of the satellite: transmission halfway around the Earth20,000 kmrequires at least 0.067 seconds.9) Moreover, if some of the satellites are on highly inclined orbits, observers at high latitudes can see the satellites at high elevation angles, which reduces interference with the signals by buildings and other objects. These qualities make LEO orbits useful for personal communications systems. The disadvantage of using LEO satellites for this purpose is that the network requires many satellites. Recall that any observer sees a satellite passing overhead for roughly 10 minutes out of its 90-minute orbit, so nine satellites would be required to provide continuous coverage of a single band on the Earth around its ground track (for an orbital altitude of 500 km, the width of this band is roughly 3,000 km, as Table 5.1 shows). For broader coverage, considerably more satellites would be needed. For example, the Iridium constellation, which is used for a variety of military and commercial purposes, has 66 satellites distributed in six different orbits with an altitude of 780 km. The six orbits are in six different orbital planes, each at an inclination angle of 86.4.10 Low earth orbits may also be useful for missions that do not require realtime communication. Such missions may need only one or a few satellites. For example, data need not be sent to ground users immediately, but can be stored and then forwarded when the satellite passes over the ground station (this arrangement is known as store-and-forward). For missions that are not time critical, the motion of the LEO satellites relative to the Earth means that a single satellite in polar orbit can cover the entire Earth. If the orbital period is chosen so that the ground coverage areas on successive orbits lie next to each other, a satellite in a polar orbit can see any spot on Earth twice a day. Some missions require low orbits. Earth observation and reconnaissance satellites intended to take high-resolution images of the Earth must be close to the Earth to get such resolution (see discussion in Appendix B to Section 11). For example, the U.S. Keyhole satellites, which took optical photographs for intelligence purposes, were usually deployed in elliptical orbits with an apogee and perigee at 1,000 and 300 km, respectively.11 These have been

9. The relay process, which requires each relay satellite to receive and retransmit the signal, also adds to the transmission time. 10. Iridium Satellite Solutions, http://www.iridium.com, accessed January 15, 2005. 11. Federation of American Scientists, KH-11 Kennan/Crystal Satellites, September 9, 2000, http://www.fas.org/spp/military/program/imint/kh-11.htm, accessed January 15, 2005.

T Y P E S O F O R B I T S , O R W H Y S AT E L L I T E S A R E W H E R E T H E Y A R E

41

replaced by a new generation of imaging satellites in similar orbits. Since these satellites move with respect to the Earth, they cannot offer continuous coverage of a particular area.

Circular Medium Earth Orbits

Satellites in circular medium earth orbits (MEO), also termed intermediate circular orbits (ICO), have altitudes between those of low earth orbits and geosynchronous orbits: from roughly 1,500 to 36,000 km. A common orbit is one with an altitude of roughly 10,000 km and an orbital period of about 6 hours. Continuous worldwide real-time coverage can be obtained with fewer satellites than are needed for a constellation of satellites in low earth orbits. For example, the ICO communications satellite system under construction will consist of 10 satellites in 2 orbits at an altitude of 10,390 km. The two orbital planes will be at 45 inclination, rotated 180 around the Earths axis with respect to one another.12 Satellites in such medium earth orbits are relatively slow moving as seen from the Earth, thus requiring fewer and simpler handover arrangements than a LEO system. The round-trip transmission time to these satellites from the ground is longer than to a satellite in low earth orbit: the ICO transmission time is 0.069 seconds, whereas for the Iridium system it is 0.0052 seconds. This longer transmission time is less of an issue for communications over long distances (a signal traveling halfway around the world would along the Earths surface require a minimum of 0.067 seconds, comparable to the time it takes for a round trip to the ICO satellite), and using higher altitude satellites reduces the number of satellites the signal must be relayed between to cover long distances. However, satellites in MEO orbits must employ radiation-hardened components (particularly to protect their computer systems) to survive long term. A special type of medium earth orbit is the semisynchronous orbit, which has a period of 12 hours and an altitude of roughly 20,000 km. Both the U.S. Navstar Global Positioning System (GPS) and Russian Glonass navigational satellites use these orbits. A navigational system needs at least four satellites within view of the user at all times, where a continuous communications system needs only one. Thus, a navigational system requires more satellites than does a communications system deployed at the same altitude: both GPS and Glonass (when fully deployed) use 24 satellites. The GPS satellites are in six orbital planes at an inclination angle of 55; Glonass is designed to use three orbital planes at an inclination angle of 65.

Molniya Orbits
Molniya orbits are highly elliptical, with a period of 12 hours and an inclination of 63.4. At this inclination, the apogee remains over the same latitude in

12. ICO Satellite Wireless Services, http://www.ico.com, accessed January 15, 2005, and Lloyds Satellite Constellations, http://www.ee.surrey.ac.uk/Personal/L.Wood/constellations/ ico.html, accessed January 15, 2005.

42

T H E P H Y S I C S O F S PA C E S E C U R I T Y

the northern (or southern) hemisphere, rather than precessing. The Soviet Union first used this type of orbit for its Molniya satellite system, hence the name. They are sometimes referred to as highly elliptical orbits (HEO). A satellite in a highly elliptical orbit with the apogee over the northern hemisphere covers Earths high-latitude regions for a large fraction of its orbital period. As discussed in Section 4, the speed of a satellite is not constant on an elliptical orbit. The satellite has a high speed as it traverses the orbit near perigee and moves slowly near apogeethus spending most of its time in the sky over the northern hemisphere. The Russian Molniya satellites are in orbits with an apogee of roughly 40,000 km and a perigee of roughly 1,000 km (or an eccentricity of 0.75). For eight of their 12-hour periods, each satellite remains visible to the regions under the apogee, with elevation angles above 70. A constellation of three satellites, with their major axes oriented at 120 with respect to each other, ensures continuous coverage of this area. Molniya orbits are also used by U.S. intelligence satellites that monitor Russia and by Russian early warning satellites that watch for U.S. missile launches.

Tundra Orbits
Like Molniya orbits, Tundra orbits have an inclination of 63.4, so their apogees remain over one hemisphere. They are typically used to provide coverage of high latitude areas, with their apogee over the northern hemisphere. However, they are not as highly elliptical as are the Molniya orbits, and their period is 24 hours rather than 12. Satellites in Tundra orbits are visible to the regions under the apogee for 12 of their 24-hour periods. Thus, it is possible to obtain continuous coverage of this region with only two satellites whose orbits are rotated 180 with respect to each other. The Russian Tundra system uses two satellites in orbits with an apogee and perigee of roughly 54,000 and 18,000 km, respectively.

Geostationary Orbits
Geosynchronous orbits have a period equal to the Earths rotation period. The most useful geosynchronous orbit is the geostationary orbit, which is a circular orbit at an altitude of 35,786 km in the equatorial plane. Because a geostationary satellite appears as a fixed point in the sky to all observers on the ground, users need no tracking equipment to send or receive signals from the satellite. Three satellites can provide worldwide coverage, excluding the polar regions. The area of visibility of the satellite is large; it is not quite half the Earthabout 43% coverage. Thus, geostationary satellites can provide continuous service over a wide geographical area. This is very useful for television and radio broadcasting, since it permits real-time data transfer over a wide geographic area without using a store-and-forward scheme. It also provides the necessary flexibility for commercial and military communications, which need to support users from widely different, nonpredetermined locations.

T Y P E S O F O R B I T S , O R W H Y S AT E L L I T E S A R E W H E R E T H E Y A R E

43

Geostationary satellites operate outside the densest regions of the Van Allen belt, but they are subject to infrequent bursts of high-energy particles from the Sun that can damage or degrade them.

Sun-Synchronous Orbits
Satellites in sun-synchronous orbits pass over a given part of the Earth at roughly the same local time of day (though not necessarily every day). That is, whenever the satellite observes a given ground location, the Sun is always in the same location in the sky. Such orbits are particularly useful for missions that take images of the Earth, because shadows from objects at a given location on the Earths surface are always cast from the same angle. This simplifies the comparison of images taken on different days to detect changes. Satellites in these orbits are often placed at low altitudes (with short periods) so that they provide complete coverage of the Earths surface at least once per day. The inclination of sun-synchronous orbits is chosen so that the precession of the orbital plane around the Earth due to gravitational irregularities keeps the plane at a constant angle with respect to a line between the Earth and Sun throughout the year. The precise inclination that produces this effect depends on the orbits altitude and eccentricity; it is typically 9698, making the orbits slightly retrograde. Figure 5.6 illustrates how a nonprecessing orbit differs from an orbit that precesses synchronously with the Sun.

Figure 5.6. Both panels show the Earth at four positions in its yearly orbit around the Sun, and the orbital plane of the same satellite in each case. Panel A shows a case in which the satellites orbit does not precess and remains in a fixed orientation with respect to space. Thus, a satellite that is directly above a location on the Earth when the local time is midnight and noon, would four months later observe this location when the local time is 6 am and 6 pm. Panel B shows a sun-synchronous orbit. The orbit is in a plane chosen to precess at a rate synchronized with the Earths trip around the Sun, so that the plane maintains a constant angle throughout the year with respect to a line between the Earth and Sun. As a result, during the entire year, this satellite observes a point on the Earth at the same local time.

44

T H E P H Y S I C S O F S PA C E S E C U R I T Y

In a special sun-synchronous orbit, called a dawn-to-dusk orbit, the satellites orbital plane coincides with the plane that divides the half of the Earth that is illuminated by the Sun from the half that is dark. If the plane were aligned slightly differently, the satellite would spend half of its time in full sunlight and half in shadow, but a dawn-to-dusk orbit allows the satellite to always have its solar panels illuminated by the Sun. For example, the Canadian Radarsat Earth observation satellites13 use such a dawn-to-dusk orbit to keep their solar panels facing the Sun almost constantly, so they can rely primarily on solar power and not on batteries.

Lagrange Points
There are five special orbits in which satellites orbit not the Earth but the Sun, and do so in a way that they maintain a fixed position relative to the Earth as it orbits the Sun. These fixed locations are called Lagrange points; there are five such points, one corresponding to each of the five orbits (see Figures 5.7 and 5.8). A satellite orbiting the Sun closer than the Earth does has a shorter orbital period than the Earths. However, such a satellite is pulled by the Earths gravitational field as well as by that of the Sun. This effect is negligible if the satellite is far from the Earth, but must be taken into account for a satellite close to Earth. For a satellite directly between the Earth and Sun, the direction of the Earths pull is exactly opposite that from the Sun, effectively canceling some of the Suns gravitational pull. At the first Lagrange point14 (L1), the net gravitational force on the satellite is the same as the Suns gravitational force on the Earth, so that the satellite orbits the Sun with the same orbital period as the Earth. A satellite in this position stays with the Earth throughout its journey around the Sun. The L1 point is about four times more distant from the Earth than the Moon is. The L1 point is particularly useful for scientific missions that study the Sun, and satellites positioned there can give early warning of increased solar winds. A second Lagrange point is located the same distance from the Earth but on the other side, directly away from the Sun. In this case, the Earths gravitational pull adds to that of the Sun, increasing the orbital speed required for the satellite to stay in orbit. In this case, the satellite keeps up with the Earth in its orbit, while it would normally fall behind. Scientific missions are positioned there as well, allowing the satellite to be maximally far from the Earth (to minimize interference), but maintain constant contact. NASA plans to place its Next Generation Space Telescope (NGST), the successor to the orbiting Hubble telescope, at or near L 2.

13. For information on Radarsat, see the Radarsat International website http://www.radarsat2.info/, accessed January 15, 2005. 14. There are analogous Lagrange points for the Earth-Moon system. These points are near the Moon and stationary with respect to it. The Lagrange points discussed here are all in the Earth-Sun system.

T Y P E S O F O R B I T S , O R W H Y S AT E L L I T E S A R E W H E R E T H E Y A R E

45

Figure 5.7. The L 1 and L 2 Lagrange points for the Earth-Sun system.

Figure 5.8. The L 4 and L 5 Lagrange points for the Earth-Sun system.

The L3 point, which lies on the other side of the Sun, directly opposite the Earth, is not very useful for satellites. The L4 and L5 points are along the Earths orbit, but precede and lag it. They are 60 away from the Earth-Sun line. Some suggest the L2 will be strategically interesting for space exploration or for space militarization.15 Since craft at L2 are in a stable position and need little fuel to remain there for extended periods of time, L2 could be used as a place to assemble other spacecraft from parts lifted piece by piece. Such a scheme could be more energy efficient than trying to assemble large structures on the Moon and more feasible than assembling them on Earth and then launching them. Objects at L2 are also out of easy observation by the Earth, being quite distant.

15. For example, see James Oberg, Will Chinas Space Plan Skip the Moon? Space News, May 24, 2004, 13.

46

T H E P H Y S I C S O F S PA C E S E C U R I T Y

Section 5 Appendix: Details of Elevation Angle and Ground Area

E L E VAT I O N A N G L E

The elevation angle of a satellite as seen by an observer is16

(5.1) where Re is the radius of the Earth, h is the altitude of the satellite, and is defined as (5.2) where l and are the latitude and longitude of the observer, respectively, and and are the latitude and longitude of the satellite. For a satellite in an equatorial circular orbit, = 0. The maximum elevation angle max seen by the observer occurs at the point in the orbit for which the longitude of the observer equals that of the satellite ( = ), so the above equation simplifies to (5.3) Thus, the maximum elevation angle max of a satellite in an equatorial circular orbit as seen by an observer at latitude l is

(5.4) For an observer at a latitude of 45, max is 17 for h = 500 km and 38 for h = 36,000 km. Recall that tall buildings can interfere with satellite reception for elevation angles of up to 70. Using this equation, for a satellite in geostationary orbit (with h = 36,000 km) this corresponds to an observer at a latitude of roughly 18.
GROUND AREA VIEWED BY A SATELLITE

From Figure 5.9, it can be shown that the radius Rarea of the maximum circular region (as measured along the Earths surface) that can be viewed by a satellite at altitude h is equal to

( )

(5.5)

16. Grard Maral and Michel Bousquet, Satellite Communications Systems, 4th ed. (West Sussex, England: Wiley, 2002), 44-45.

T Y P E S O F O R B I T S , O R W H Y S AT E L L I T E S A R E W H E R E T H E Y A R E

47

where Re is the radius of the Earth and the angle is in radians. The fraction F of the Earths surface this region represents is (5.6)
Figure 5.9. This figure shows the geometry used to calculate the ground area visible to a satellite at altitude

h.

If the minimum elevation angle min at which the user can communicate with the satellite is greater than 0, then by using the law of sines, it can also be shown, from Figure 5.9, that the radius Reff of the effective observable region (as measured along the Earths surface) is

(
( )

))

(5.7)

where the angles are expressed in radians. The fraction of the Earths surface covered by this region is

(5.8) where

(5.9)

48

T H E P H Y S I C S O F S PA C E S E C U R I T Y